Regulation of BRCA1 stability through the tandem UBX domains of isoleucyl-tRNA synthetase 1

Aminoacyl-tRNA synthetases (ARSs) have evolved to acquire various additional domains. These domains allow ARSs to communicate with other cellular proteins in order to promote non-translational functions. Vertebrate cytoplasmic isoleucyl-tRNA synthetases (IARS1s) have an uncharacterized unique domain, UNE-I. Here, we present the crystal structure of the chicken IARS1 UNE-I complexed with glutamyl-tRNA synthetase 1 (EARS1). UNE-I consists of tandem ubiquitin regulatory X (UBX) domains that interact with a distinct hairpin loop on EARS1 and protect its neighboring proteins in the multi-synthetase complex from degradation. Phosphomimetic mutation of the two serine residues in the hairpin loop releases IARS1 from the complex. IARS1 interacts with BRCA1 in the nucleus, regulates its stability by inhibiting ubiquitylation via the UBX domains, and controls DNA repair function.

The secondary structure is displayed at the top and bottom of the sequence. c A surface representation of the UBX domains with the degree of sequence conservation denoted by the same colors as in b.
c Protein extracts from HeLa cells transfected with control siRNA or two siRNAs IARS1 were immunoblotted using anti-IARS1, anti-CtIP, anti-EXO1, and anti-tubulin (loading control) antibodies. d U2OS cells overexpressing GFP-EXO1 were transfected with control siRNA or two siRNAs IARS1 and exposed to laser microirradiation. Laser stripes were examined at the indicated time point. EXO1 recruitment to DSB site was calculated from GFP positive cells over total counted cells. c, d The data are representative of three independent experiments.
Source data are provided as a Source Data file.

Supplementary Note
Overall structure of EARS1 GgEARS1 forms an elongated shape with the N-terminal region formed by the acceptor end- Additional disordered regions include residues 545-546 and 560-563. However, most of the disordered regions become ordered upon complex formation with IARS1.

Structural comparison with archaeal EARS1 and bacterial and mammalian QARS1s
The overall structure of GgEARS1 is highly similar to those of QARS1 and archaeal EARS1, and can be superimposed on those of human QARS1 (PDB 4R3Z) 2 , E. coli QARS1 (PDB 1QTQ) 3 , and archaeal EARS1 (PDB 3AII) 1 with the rms deviation values of 1.8-2.5 Å (386-430 Cα atoms). In the superimposed structures, eukaryotic EARS1 more resembles bacterial and eukaryotic QARS1s than archaeal EARS1 ( Supplementary Fig. 3). In all five domains of QARS1 and EARS1, the Dβ domain exhibits the largest deviations, and the selectivity for anticodon binding (Glu or Gln) could be determined by these differences ( Supplementary   Fig. 3). Three regions of GgEARS1, namely, the IARS1-binding site and anticodon-and acceptor end-binding sites, exhibit most notable differences with respect to QARS1s and archaeal EARS1.

Four loops provide specificity for the binding of tRNA Glu to vertebrate EARS1
The most significant difference between GgEARS1 and archaeal EARS1 or QARS1 is observed in the opposite face of the tRNA-binding interface ( Fig. 1 and Supplementary Fig.   3). The β-hairpin loop (residues 684-695) in the Pβ domain is present only in vertebrate EARS1, not in archaeal EARS1 or bacterial or mammalian QARS1. In the apo GgEARS1 structure, this loop is disordered, but in the IARS1-bound state, several main chains of the hairpin loop pair with each other to form a well-ordered β-hairpin structure.
Notable differences are also observed at the U33 and C34 site: the S17-S18 loop in the Dβ domain is present only in GgEARS1 (light blue), not in archaeal EARS1 or bacterial QARS1. The HsQARS1 is more similar to GgEARS1 in this region. Nevertheless, the corresponding loop of human QARS1 is 6 to 10 Å away from the S17-S18 loop of

Determination of the EARS1 structure
The GgEARS1 structure was determined by a combination of the molecular replacement and single-wavelength anomalous dispersion methods using crystals containing Hg-derivatized

Data collection using XFELs
The fixed-target serial femtosecond crystallography (FT-SFX) experiment using a nylon mesh and X-ray pulses was conducted at the Nano Crystallography and Coherence Imaging experimental hutch of the Pohang Advanced Light Source X-ray free electron laser 9,10 (PAL-XFEL). The X-ray energy was 9.7 keV (1.2782 Å) and the photon flux was ~5 × 10 11 photons per pulse (<20 fs duration). The X-ray pulse was focused to 5 (horizontal) × 8 (vertical) μm 2 (FWHM) using a Kirkpatrick-Baez mirror 11 . The data were collected from a sample chamber filled with helium gas at ambient pressure and room temperature (RT) and were recorded using a MX225-HS detector (Rayonix) with a 4 × 4 binning mode (pixel size: 156 × 156 µm 2 ). The motion stage for FT-SFX was designed to allow raster scanning of beams up to 60 Hz provided by PAL-XFEL and was custom-built by SmartAct. Piezo

SFX data processing
The diffraction pattern was monitored using OnDA 12 , and the hit images, defined as those images containing a minimum of 15 peaks, were filtered using the Cheetah program 13 . The 'peak-finder 8' algorithm was used with key peak-finding parameters, and yielded an average hit-rate of 69.1% (301,437 hits) for GgEARS1-IARS1 crystals. The diffraction images were indexed, integrated, merged, and post-refined using CrystFEL 14 . Indexing was performed using a combination of MOSFLM 15 , which uses prior unit cell information for indexing, and DirAx. The detector geometry was refined with several iterations of the detector geometry optimization using CrystFEL's geoptimiser for increasing the indexing rate, and the peaks used for indexing were those outputted from Cheetah 13 . Preliminary indexing results, which were obtained in synchrotron, strongly suggested a primitive orthorhombic lattice, and a subsequent round of indexing was performed. The final indexing rate was 16.9% (50,804 indexed patterns) for Gg EARS1-IARS1. The intensities were merged using scaling, partiality, and post-refinement with a partialator in CrystFEL 14 .

RNA extraction and cDNA synthesis
siRNA-transfected cells were lysed with 500 µL TRIzol, and total RNA was extracted according to the manufacturer's instructions (Invitrogen). In brief, 100 µL chloroform was added to lysed Hela cells, samples were centrifuged at 17,000 g for 15 min at 4°C, 150 µL aqueous phase was collected, and RNA was precipitated from the aqueous phase by mixing with an equal volume of isopropanol. After incubation for 10 min, samples were centrifuged at 17,000 g for 15 min at 4°C. The RNA pellet was washed twice with 75% (v/v) ethanol, briefly air-dried for 5 min, and dissolved in diethylpyrocarbonate-treated water. The purity and yield of RNA were determined using a NanoDrop 2000 spectrophotometer (Thermo Scientific). For cDNA synthesis, 1 µg total RNA was reverse-transcribed using oligo-dT and the ImProm-II Reverse Transcription System (Promega) according to the manufacturer's instructions. The RT-PCR conditions were one cycle of 25°C for 5 min, 42°C for 60 min, and 72°C for 7 min. For RT-qPCR analysis, each cDNA was diluted 2-fold with nuclease-free water.

Quantitative real-time PCR
The mRNA level of BRCA1 was measured by quantitative real-time PCR using a StepOnePlus Real-Time PCR system (Applied Biosystems) with SYBR Premix Ex Taq (Takara) according to the manufacturer's instructions. The sequences of qPCR primers are shown in Supplementary Table 2. Data were normalized against β-actin expression, and relative expression was calculated using the ΔΔCT method 16 .

Immunofluorescence microscopy
To detect the subcellular localization of IARS1, transfected HEK293T cells were seeded on To analyze Rad51 and γH2AX focus formation, IARS1-depleted U2OS cells were transferred to LabTek™ 4-well chamber slides at a density of 7 × 10 4 cells per well. They were exposed to ionic radiation using a RS 2000 irradiator (Rad Source) at a dose rate of 10 Gy min -1 . Cells were pre-extracted with CSK buffer (10 mM PIPES, 100 mM NaCl, 300 mM sucrose, 3 mM MgCl2, 1 mM EGTA, and 0.5% Triton X-100) for 10 min on ice and fixed with 4% paraformaldehyde for 20 min at RT. After blocking for 30 min at RT, cells were incubated with the indicated antibodies diluted in blocking buffer overnight at 4°C. Rabbit anti-Rad51 (8875S, Cell Signaling, 1:1,000) and goat anti-γH2AX (05-636, Millipore, 1:1,000) antibodies were used. After three washes with PBS containing 0.05% Triton X-100, cells were incubated with an Alexa Fluor 488-conjugated secondary antibody for 30 min and mounted using ProLong ® Gold antifade reagent (Vector Laboratories). Confocal images were acquired with an LSM880 confocal microscope (Carl Zeiss). Image acquisition and analysis were performed with ZEN blue software (Carl Zeiss).

Laser microirradiation
A total of 3 × 10 5 U2OS cells were plated in confocal dishes (SPL), incubated for 1 day, and transfected with 2 µg of a GFP-EXO1-expressing plasmid using Lipofectamine 3000 according to the manufacturer's instructions. After 4 hr, media were replaced by media containing 10 µM 5-bromo-2'-deoxyuridine. The following day, laser microirradiation was performed using a 355 nm ultraviolet A laser, and cells were incubated in a 37℃ chamber containing 5% CO2. After each laser microirradiation, images of cells were obtained every 10 sec for 5 min using a LSM880 confocal microscope. The intensity of each laser strip was determined using Zen blue software, and the values were normalized against the baseline values. At least ten cells were used for quantification.